1902.] STENGEL— SPECIFIC PRECIPITINS. 409 



animal body, generate or cause the generation of substances which 

 in a more or less specific manner precipitate the substance or com- 

 pound used in the inoculations. This was first demonstrated in the 

 case of milk by Bordet, who found that when he injected rabbits 

 with sterilized milk and repeated the injection at short intervals 

 the blood serum of the animals subsequently caused precipitation 

 of the milk, while normal serum had no such action. Later it was 

 found by other experimentors that this precipitation is more or less 

 specific, and the serum of a rabbit treated with cow's milk contains 

 a precipitin for cow's milk and not for the milk of other animals, 

 while the serum of animals treated with goat's milk is similarly 

 specific. The same experiment was performed with human milk 

 and the result was similar. The method of carrying out these ex- 

 periments was as follows : The rabbits were inoculated subcutane- 

 ously or intraperitoneally at intervals of several days with sterilized 

 milk, the quantity varying between lo and 50 c.c. The steriliza- 

 tion was accomplished by heating for an hour at 65° C. or by the 

 use of chloroform. Other rabbits were treated with milk of differ- 

 ent sorts as controls. After the treatmenc had been repeated for a 

 number of times and each rabbit had received about 100 c.c. of 

 milk the serum was obtained by bleeding the rabbit and allowing 

 the blood to coagulate. The serum diluted with four or five times 

 its bulk of water was mixed with milk diluted i to 40 and the mix- 

 ture allowed to stand for some hours. The precipitation was then 

 observed in the milk corresponding with that with which the 

 animal furnishing the serum had been treated. 



A very similar series of experiments has been performed with egg 

 albumin. Repeated injections of raw egg albumin cause the devel- 

 opment in the blood of a more or less specific precipitin. When 

 crystallized egg albumin was used a specific precipitin was devel- 

 oped. This precipitates the egg albumin, but not globulin, and on 

 the other hand, when serum globulin is used in the injections, the 

 precipitin has no effect upon egg albumin. In the case of these 

 more narrowly specific tests the results have been somewhat at 

 variance. Thus it was found that the serum of animals injected 

 with globulin obtained from bullock's serum had some effect upon 

 blood corpuscles of fowl and also an effect upon sheep globulin. 

 Other instances of non-specific action might be cited, but this one 

 reference is sufficiently illustrative of all. 



My own experiments in conjunction with Dr. C. Y. White in 



